CN113102269A - Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof - Google Patents
Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof Download PDFInfo
- Publication number
- CN113102269A CN113102269A CN202110372277.7A CN202110372277A CN113102269A CN 113102269 A CN113102269 A CN 113102269A CN 202110372277 A CN202110372277 A CN 202110372277A CN 113102269 A CN113102269 A CN 113102269A
- Authority
- CN
- China
- Prior art keywords
- component
- prefabricated
- quality
- module
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
A prefabricated part finished product quality detection system based on three-dimensional stereo vision and an application thereof relate to the field of prefabricated part quality detection, a master control module is respectively connected with an image acquisition module, an image processing module, a transportation tool parking sensing module, a component space sensing module and a network communication module, the master control module is responsible for comprehensive control of all functional modules in the prefabricated part finished product appearance quality detection system based on three-dimensional vision, information interaction and command control of an integral hardware platform, a hardware part of the image acquisition module mainly comprises a binocular vision camera and a video transmission line of component quality detection points, the image processing module detects prefabricated part model matching and prefabricated part appearance damage, the component space sensing module comprises an infrared detector, and whether a prefabricated part to be quality detected reaches the component quality detection points is monitored. Not only provides accurate and efficient quality inspection detection means, but also provides basis for inventory logistics of prefabricated part production enterprises and decision judgment of construction site engineering management personnel.
Description
Technical Field
The invention relates to the field of prefabricated part quality detection, in particular to a prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof.
Background
The assembly type building is beginning to be widely applied to a plurality of building fields such as industrial buildings, roads and bridges, hydraulic engineering and the like at the beginning of the 20 th century. After 90 s, the research and application of the assembly type building technology in China began to be started and rapidly developed. Compared with the traditional cast-in-place concrete building, the core innovation of the prefabricated building is that the main body of the building adopts a modularized assembled structure, and the concrete prefabricated part is manufactured in a production mode of batch manufacturing in a workshop. The prefabricated components such as wall boards produced on a production line are mainly influenced by the size of a mould table, the mould table has various specifications such as 3.5 mx 9m, 4 mx 9m, 3.5 mx 12m, 4 mx 12m and the like, the size of the corresponding components is smaller than or equal to the size, so the prefabricated components belong to large-size industrial products, meanwhile, the prefabricated components are usually composed of concrete and reinforcing steel bars, the mass of the prefabricated components is larger, the energy consumption and the transportation cost in the transportation process cannot be ignored, the components are not influenced by more production cost, production period and other factors than other industrial products in the production and use processes, and the prefabricated components are difficult to be transported and turned over as fast and conveniently as other products due to the larger size of the prefabricated components, and the mechanism design of a conventional machine vision detection means and inspection equipment cannot meet the quality detection requirement of the prefabricated components. Although the prefabricated part structure model is a single building assembly obtained by disassembling a drawing on the premise of following a certain standard specification through a building structure drawing, the size and the style of the prefabricated part are different due to the influence of factors such as a building structure, a use function, pipeline wiring and the like, the corresponding prefabricated parts cannot be replaced mutually in the assembly process of construction and construction, and the prefabricated parts need to pass through a plurality of links such as design, production, storage, logistics and the like in an industrial chain before being assembled and used on a construction site, a certain time cost is required in the process, if quality problems are found in the installation process of the prefabricated parts when the prefabricated parts are conveyed to the construction site, broken parts and wrong parts cannot be replaced in time within a certain time range, the construction progress is seriously influenced, and the whole construction cost is improved.
The finished prefabricated components are moved into a storage place, and the finished prefabricated components are stored and placed in a transverse and vertical mode, the transverse components (such as beams, laminated plates, double T-shaped plates and the like) need to be placed on the flat and hardened ground or stacked by steel frames, the vertical components (such as prefabricated walls and plates) need to be placed vertically, and the prefabricated components are stacked by professional wallboard stacking racks or A-shaped racks. It is difficult to carry out convenient comprehensive quality testing under two kinds of storage conditions of above-mentioned prefabricated component to current many enterprises still rely on artifical vision to carry out the outward appearance according to the design drawing to the quality testing link of component and detect, technical staff's detection working strength is great, and has certain safe risk, consequently, the intelligent quality testing means of efficient that need introduce urgently improves quality inspection efficiency and practices thrift the cost of labor.
The prefabricated part has large volume and mass, so that the transportation cost is high, the quality problems of damage, stress deformation and the like easily occur in the processes of repeated hoisting and long-distance transportation of the prefabricated part, the working strength and difficulty of detection technicians are increased by repeated detection, and the problems of wrong detection, missing detection, non-uniform detection means and detection standards are easily caused.
Therefore, a system for detecting the quality of the finished prefabricated part based on three-dimensional stereoscopic vision is designed, and an effective detection means is provided for the field of building quality detection.
Disclosure of Invention
The invention aims to provide a prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof, and aims to solve the problems of long debugging time, error detection, omission and the like in the prior art. The prefabricated part finished product quality detection system based on three-dimensional stereo vision is an independent system and can be applied to other systems as a subsystem.
The technical scheme of the invention is as follows: the utility model provides a prefabricated component finished product quality detecting system based on three-dimensional visual sensation, including total control module, the image acquisition module, image processing module, the perception module is berthhed to the transport means, component space perception module and network communication module, total control module respectively with the image acquisition module, the image processing module, the perception module is berthed to the transport means, component space perception module and network communication module are connected, total control module is responsible for the comprehensive management and control of each functional module in the prefabricated component finished product appearance quality detecting system based on three-dimensional visual sensation, information interaction and the command control of whole hardware platform.
The hardware part of the image acquisition module mainly comprises two groups of binocular vision cameras and video transmission lines of component quality detection points (the pixels of the acquired images of the components are large, and the transmission rate is not high due to the adoption of a traditional wireless transmission mode, so that a wired mode is adopted), and the image acquisition module has the functions of acquiring three-dimensional image information of the appearance of the prefabricated components and uploading the acquired image information of the components to the image processing module.
The image processing module mainly realizes two functions of prefabricated part model matching detection and prefabricated part crack damage detection, and can judge whether the detected prefabricated part is a received component transfer order inner part or not through the prefabricated part image information acquired by the image acquisition module after passing through the image processing module, and further detect whether the appearance of the prefabricated part has quality problems such as crack damage and the like.
The vehicle stop sensing module comprises a vehicle stop sensing instrument and is used for monitoring whether the vehicle reaches a monitoring point.
The component space sensing module monitors whether the prefabricated component to be subjected to quality inspection reaches a component quality detection point.
The network communication module is used for performing information interaction between modules and with the air suspension system in the three-dimensional stereoscopic vision prefabricated part finished product quality detection system, and when the component quality detection system is used as a subsystem of the component storage and transportation system, the network communication module is also used for performing information interaction with the component comprehensive control subsystem, the storage air suspension subsystem and the component transportation subsystem.
The master control module realizes the detection of the spatial position of the member and the parking position of the vehicle by controlling the spatial sensing module of the member and the parking sensing module of the transport tool; the appearance three-dimensional image information of the prefabricated part is obtained by controlling the image acquisition module, and the image processing module can be controlled to realize the matching detection and the crack damage detection of the prefabricated part; the information interaction among all functional modules in the system is realized by controlling the network communication module; when the quality inspection system is used as a subsystem of the component warehousing and transportation system, the master control module is also responsible for realizing information interaction with the upper computer management system through the control network communication module and receiving tasks of transportation orders issued by the upper computer management system, so that the function of comprehensively managing and controlling the quality inspection system is realized.
The method comprises the following steps of:
And 2, if the system self-checking is normal, the master control module controls the protective cover of the binocular vision camera to be opened, namely the protective cover is opened by the binocular vision camera pre-embedded on the ground below the component quality detection point, so that the binocular vision camera can normally acquire three-dimensional image information of the appearance of the prefabricated component, and the protective cover is in a closed state when the detection work is not carried out at ordinary times, so that a group of binocular vision cameras pre-embedded on the ground are protected. The quality inspection system receives component transfer orders through the network communication module (the transfer orders can manually input transfer order information; or when the quality inspection system is used as a subsystem of a warehousing transfer system, the quality inspection system receives the transfer order information of the component comprehensive control subsystem).
And 3, monitoring whether the component transfer tool reaches a detection point by the transport tool stop sensing module. If the component transferring tool does not arrive within the set waiting time, the quality inspection system can continuously detect whether the transferring tool arrives within the polling detection time which is not more than the set waiting time, if the transferring tool does not arrive all the time, the quality inspection system enters a dormant state, and then transmits a dormant information instruction through the network communication module to send a dormant information prompt.
And 4, if the component transfer tool arrives within the specified time, the component space sensing module immediately starts working to detect whether the prefabricated component reaches the detection range, if the prefabricated component does not arrive within the specified waiting time, the quality inspection system can also continuously detect whether the component arrives within the polling detection time which is not more than the set waiting time, if the prefabricated component does not arrive all the time, the system enters a dormant state, and meanwhile, a dormant information instruction is transmitted through the network communication module to send a dormant information prompt.
And 5, if the component is determined to reach the detection range, controlling the image acquisition module to start working by the master control module, namely, controlling the upper and lower groups of binocular vision cameras at the component quality detection point to start working, and acquiring the appearance three-dimensional image information of the prefabricated component, correcting the cameras and acquiring the upper and lower layer disparity maps of the prefabricated component. After the image information is acquired, the master control module controls the image processing module to start working, the three-dimensional image information of the appearance of the prefabricated part acquired by the image acquisition module can be matched with a structural model of the prefabricated part for detection, if the matching fails, a matching error instruction is transmitted by the network communication module, an alarm prompt is sent, then the master control module controls the network communication module to send an instruction for placing the part into a wrong part area to the navigation and hoisting system, and the navigation and hoisting system hoists and transports the mistaken part to the wrong part area.
And 6, if the component to be detected is successfully matched with the component structure model, the detected component is proved to be a component contained in the received transfer order, the master control module controls the image processing module to carry out next component crack damage detection operation, and whether the component has quality problems such as cracks, damages and the like is detected. If the image processing module judges that the quality problems such as crack damage and the like exist in the detected component, the quality inspection system transmits a detection disqualified instruction through the network communication module and sends an alarm prompt, and then the master control module controls the network communication module to send an instruction for placing the component into a bad component area to the navigation hanging system, so that the navigation hanging system can hang and transport the component with the quality problem to the bad component area.
And 7, if the components are successfully matched and the quality inspection is qualified, the master control module controls the network communication module to send an instruction for hoisting the prefabricated components to the transfer tool to the navigation crane system, and then the navigation crane system hoists the detected qualified prefabricated components to the component transfer tool.
And 8, the master control module sends a corresponding instruction according to whether the components needing to be transported in the transport order are hoisted or not. If the transfer order is not complete and hoisting continues, the system continues to detect if a new vehicle arrives, as the multiple components required for each transfer order may require two or even multiple component transfer tools to be co-transferred. And if the transfer order is finished, the quality inspection system finishes the work.
In the component quality inspection processing flow of the three-dimensional stereoscopic prefabricated component finished product appearance quality detection system, after the prefabricated component appearance three-dimensional image information is acquired, the acquired prefabricated component image information is processed, and the image processing process is divided into two parts, namely component model matching detection and component crack damage detection.
Image acquisition:
as the size of the prefabricated part is large, a single group of binocular vision cameras cannot acquire a complete appearance image of the prefabricated part in the process of transporting the prefabricated part by the aerial crane, each group of binocular vision cameras in the upper and lower positions are designed, when the prefabricated part reaches a part quality monitoring point, two groups of binocular vision cameras on the upper and lower sides of the part quality monitoring point start to work, calibration is firstly carried out, and the internal and external parameters of each group of two cameras and the pose relation of each group of two cameras are determined. After the binocular calibration is finished, two groups of binocular vision cameras acquire appearance images of the prefabricated part, wherein an upper position binocular vision camera acquires the upper surface and side surface appearance images of the prefabricated part, a lower position binocular vision camera acquires the bottom surface image of the prefabricated part, and then distortion correction is carried out on the acquired images, so that the two corrected images are located on the same plane and are parallel to each other.
And after two groups of the corrected and over-distorted prefabricated part images are obtained, performing parallax calculation on the two images by a binocular triangulation distance measuring method to obtain a parallax image. Because the prefabricated part hoisted by the gantry crane is suspended above the quality detection point of the prefabricated part, the two groups of binocular vision cameras can obtain two processed parallax images, the binocular vision cameras in the upper direction can obtain the parallax images of the side surface and the upper surface of the prefabricated part, and the binocular vision cameras in the lower direction obtain the parallax images of the bottom surface of the prefabricated part.
Model matching:
the method comprises the steps of processing a parallax map obtained in the image acquisition process, obtaining a depth map of a prefabricated part through mathematical calculation, and separating the prefabricated part from surrounding environment information by using a region growing algorithm to obtain an image only with the depth value of the prefabricated part because the prefabricated part hoisted by a crane is suspended above a quality detection point of the prefabricated part and the obtained depth map contains the prefabricated part and the surrounding environment information.
And carrying out filtering processing on the obtained prefabricated part depth map so as to obtain a more obvious edge, then carrying out characteristic point detection on the prefabricated part depth map through the steps of Gaussian blur, gradient amplitude and direction calculation, non-maximum suppression, double threshold value and lagging boundary tracking, and carrying out contour extraction on the detected result so as to obtain the contour map of the prefabricated part which is detected in the prefabricated part quality detection point.
And comparing the profile obtained by processing the depth map of the prefabricated part with a prefabricated part model, wherein the profile characteristics of each prefabricated part can be extracted because the upper and lower positions of the quality monitoring point of the prefabricated part are provided with a group of binocular cameras, and the prefabricated part model also comprises the profile of the upper surface and the bottom surface of the prefabricated part. Therefore, in the comparison process, the upper outline and the lower outline are compared twice, so that the matching result of the detected prefabricated part and the prefabricated part model is more accurate, and the situation that the prefabricated part has the same base outline or the same upper surface outline to generate misjudgment is prevented.
And when the model matching detection result shows that the component is the component contained in the transfer order, the three-dimensional visual prefabricated component finished product appearance quality detection system starts to perform crack damage detection.
Firstly, carrying out coordinate transformation on the obtained prefabricated part depth map to obtain point cloud data, and then carrying out three-dimensional reconstruction. The prefabricated part has a large volume, so that the point cloud data of the prefabricated part is huge, the calculation time is increased during calculation of a point cloud fitting plane, the calculation load of a computer is increased, the detection process is slowed to a certain extent, and the surface structure of the prefabricated part is almost the combination of a large-area plane and a protruded edge and corner surface under the common condition, so that the point cloud data are simplified by using a method of combining curvature sampling and random sampling, redundant point clouds are reduced, and the appearance damage detection process is accelerated under the condition of ensuring that the point cloud data of the prefabricated part is relatively complete.
The obtained point cloud data set (N) reserves the curved surface characteristics of the surface of the prefabricated part, then a best fit point cloud plane (P) of the surface of the prefabricated part is obtained through an RANSAC algorithm, the obtained point cloud data set (N) and the obtained best fit point cloud plane (P) are aligned by an ICP point cloud registration method, after the alignment operation is carried out, the best fit point cloud plane (P) and the point cloud data set (N) are in the same coordinate system, then an x-y-Z coordinate system is established by the best fit point cloud plane (P), and the fitting plane (P) is equivalent to a standard plane with Z =0 in space compared with the point cloud data set (N). The appearance damage information of the prefabricated part can be judged through the coordinate value of the point cloud set (N) Z, when the space point Z coordinate in the point cloud data set (N) is a negative value, the point is concave, and when the space point Z coordinate is a positive value, the point is convex, and the depth and the height of the concave or the convex can be obtained through calculating the curvature. And comparing the structural model with the prefabricated part structural model to output whether the prefabricated part has a crack damage result.
The invention has the beneficial technical effects that: a prefabricated component finished product quality detection system based on three-dimensional stereoscopic vision and application thereof can provide an accurate and efficient quality detection means, meanwhile, through close coupling with other systems, quality detection information of components is reported in time, the transfer state of the components is tracked, a basis is provided for inventory logistics of prefabricated component production enterprises and decision judgment of construction site engineering managers, powerful support is provided for constructing a complete assembly type building integrated platform, and the intelligent level of the building prefabricated component quality detection technology is improved.
According to the invention, two groups of binocular vision cameras arranged at the quality detection points of the components are used for acquiring the three-dimensional image information of the appearance of the prefabricated components in real time, the structural model matching detection of the prefabricated components and the appearance crack damage detection of the prefabricated components are combined in the image processing process of the prefabricated components to automatically sort the components, the quality inspection automation is realized, the quality inspection result can be reported and pushed to a superior management system, and managers can conveniently and timely acquire the quality information of the prefabricated components. The invention solves the problems that the quality inspection result of the prefabricated part is not standardized and normalized in the conventional machine vision inspection means and the traditional manual quality inspection mode, carries out comprehensive and convenient identification and crack damage detection before the finished prefabricated part enters the link of logistics transportation each time, reduces the loss of time and cost caused by transferring wrong parts and parts with quality problems, reduces the waste of resources, reduces the occurrence of events seriously influencing the construction progress of the assembly type building to a certain extent, ensures the construction period and reduces the construction cost.
Drawings
FIG. 1 is a functional block diagram of a detection system of the present invention.
FIG. 2 is a flow chart of the component quality inspection process of the inspection system of the present invention.
FIG. 3 is a diagram of the operation of the detection system of the present invention as a subsystem of a warehousing and transportation system.
FIG. 4 is a flow chart of the detection system component model matching detection of the present invention.
FIG. 5 is a flow chart of the visual damage detection of the components of the detection system of the present invention.
FIG. 6 is a diagram of the physical architecture of the detection system of the present invention.
FIG. 7 is a schematic diagram of the prefabricated component warehousing transportation system of the present invention.
Detailed Description
The invention is further illustrated by the following figures and examples.
Example 1
The detection device comprises a transport tool 1, a control box 2, a transport tool parking perceptron 3, a component perception module 4 with infrared rays on two sides, a binocular vision camera 5, a suspension component 6, a wrong component area 7, a bad component area 8 and a component storage area 9, wherein a master control module, an image processing module and a network communication module are arranged in the control box 2 and used for manually controlling the start and stop of the whole quality detection system. The quality inspection system comprises a work start-stop button which is necessary when the quality inspection system and the gantry crane system are integrated to be an independent system, and also comprises hardware equipment required by the whole quality inspection system, such as a video transmission line, an image processor, a development board and a processor.
The transport means 1 loads the finished prefabricated components, and is close to the detection device, the transport means parking perceptron 3 and the component perception modules 4 with infrared rays on both sides of the transport means parking perception module of the quality inspection system detect that the transport means 1 and the components reach the quality detection range, the upper and lower groups of binocular vision cameras 5 of the quality inspection system start working, and the quality inspection system executes the quality inspection work. The detection is started by using the detection method of the prefabricated part finished product quality detection system based on the three-dimensional stereo vision, wherein the wrong part area 7 is arranged behind the bad part area 8 because when the quality detection system identifies that the detected component is not a component in the transfer order, the component in the storage aerial lifting system is lifted to the wrong part area 7 by the suspension component 6, the component in the wrong part area 7 can be continuously used, and the wrong part area 7 is arranged behind the bad part area so as to be convenient for subsequently lifting the wrong part to the component storage area 9. And the components in the damaged component area 8 are uniformly destroyed or remanufactured by a production enterprise without being hoisted to the component storage area 9.
The component quality detection points are generally arranged at the entrances and exits of storage places of production enterprises and construction sites, and can also be arranged at the loading and unloading points of the transport tool 1, wherein the component transport tool adopts a large-scale crane, the transport mode is mostly transverse hoisting, so that the crane system is not easy to generate gravity center offset when moving the prefabricated components, and in the process of moving the prefabricated components by the crane system, the transverse hoisting can keep the stable moving track of the prefabricated components as far as possible, the gravity center offset is avoided, and the component weight is prevented from damaging the stress structure of the component.
Example 2
When the three-dimensional stereoscopic vision-based prefabricated part finished product quality detection system is integrated with a navigation crane system, the detection method used as an independent system for the prefabricated part quality detection comprises the following steps:
And 2, if the system self-checking is normal, the master control module in the quality inspection system controls the protective cover of the binocular vision camera to be opened, namely the binocular vision camera pre-embedded on the ground below the component quality detection point opens the protective cover to ensure that the binocular vision camera can normally acquire three-dimensional image information of the appearance of the prefabricated component, and when the detection work is not carried out at ordinary times, the protective cover is in a closed state to protect a group of binocular vision cameras pre-embedded on the ground. The quality inspection system will then receive the component transfer order (manually entering transfer order information) via the network communication module.
And 3, after the quality inspection system receives the transfer order, the transport tool parking sensing module monitors whether the component transfer tool reaches a detection point. If the component transportation tool does not arrive within the set waiting time, the quality inspection system can continuously detect whether the transportation tool arrives within the polling detection time which is not more than the set waiting time, and if the transportation tool does not arrive all the time, the quality inspection system enters a dormant state, then transmits a dormant information instruction through the network communication module, sends a dormant information prompt and prompts field workers to enable the quality inspection system to enter the dormant state.
And 4, if the component transfer tool arrives within the specified time, the component space sensing module immediately starts to work to detect whether the prefabricated component reaches the detection range, if the prefabricated component does not arrive within the specified waiting time, the quality inspection system can also continuously detect whether the component arrives within the polling detection time which is not more than the set waiting time, if the prefabricated component does not arrive all the time, the system enters a dormant state, and meanwhile, a dormant information instruction is transmitted through the network communication module to send a dormant information prompt to prompt field workers that the quality inspection system enters the dormant state.
And 5, if the component is determined to reach the detection range, controlling the image acquisition module to start working by the master control module, namely, controlling the upper and lower groups of binocular vision cameras at the component quality detection point to start working, and acquiring the appearance three-dimensional image information of the prefabricated component, correcting the cameras and acquiring the upper and lower layer disparity maps of the prefabricated component. After the image information is acquired, the master control module controls the image processing module to start working, the three-dimensional image information of the appearance of the prefabricated part acquired by the image acquisition module can be matched and detected with a prefabricated part structural model (when the quality inspection system and the aerial lifting system are integrated to be an independent system, the prefabricated part structural model is stored in the image processing module of the quality inspection system), if the matching fails, a matching error instruction can be transmitted by the network communication module, an alarm prompt is given, a field worker is prompted about the non-transshipment order internal member of the member, then the master control module controls the network communication module to send an instruction for placing the member into a wrong part area to the aerial lifting system, and the aerial lifting system lifts and transports the wrong member to the wrong part area.
And 6, if the component to be detected is successfully matched with the component structure model, the detected component is proved to be a component contained in the received transfer order, the master control module controls the image processing module to carry out next component crack damage detection operation, and whether the component has quality problems such as cracks, damages and the like is detected. If the image processing module judges that the quality problems such as crack damage and the like exist in the detected component, the quality inspection system transmits a detection unqualified instruction through the network communication module, sends an alarm prompt to prompt field workers that the component has the quality problems such as crack or damage, and then the master control module controls the network communication module to send an instruction for placing the component into a bad component area to the navigation and hoisting system, so that the navigation and hoisting system places the component with the quality problems into the bad component area.
And 7, if the components are successfully matched and the quality inspection is qualified, the master control module controls the network communication module to send an instruction for hoisting the prefabricated components to the transfer tool to the navigation crane system, and then the navigation crane system hoists the detected qualified prefabricated components to the component transfer tool.
And 8, the master control module in the quality inspection system sends a corresponding instruction according to whether the components needing to be transported in the transportation order are hoisted or not. If the transfer order is not complete and hoisting continues, the system continues to detect if a new vehicle arrives, as the multiple components required for each transfer order may require two or even multiple component transfer tools to be co-transferred. And if the transfer order is finished, the quality inspection system finishes the quality inspection work.
When the quality detection system is integrated with the gantry crane system and used for detecting the quality of the prefabricated parts as an independent system, in order to test the quality detection effect of the prefabricated parts, transfer orders are drawn up randomly (the number of the parts required by the orders is larger than the sample capacity), and the prefabricated parts with different numbers (including wrong parts and bad parts) are adopted for testing. The sample capacity is 50 pieces, 100 pieces and 200 pieces respectively. When the sample capacity is 50 pieces (32 pieces are order internals and fracture damage exists in 5 pieces, 18 pieces are not order internals), the accuracy of component matching is 100%, and the accuracy of fracture damage detection is 100%; when the sample capacity is 100 pieces (76 pieces are order internals and 9 pieces have fracture damage, and 24 pieces are not order internals), the matching accuracy is 98%, and the fracture damage detection accuracy is 100%; when the sample volume is 200 pieces (148 pieces are order internals and 23 pieces have fracture damage, 52 pieces are not the order internals), the matching accuracy is 95%, and the fracture damage detection accuracy is 99%. As the quality inspection system does not perform the next crack damage detection once identifying the components required in the detected component non-transfer order, the crack damage detection accuracy rate is slightly higher than the matching accuracy rate. And the time required to test one component is about 20 seconds.
Examples 3,
When the prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision is used as a subsystem in a component warehousing and transferring system, the detection method is used for the quality detection of the prefabricated parts, and comprises the following steps:
in a transportation order business processing process, after a manager edits a transportation order at a mobile terminal, the order is sent to a component comprehensive control system, the component comprehensive control system can simultaneously send the transportation order to a component transportation subsystem, a storage navigation crane subsystem and a three-dimensional stereoscopic prefabricated component finished product appearance quality detection subsystem, and meanwhile, the component comprehensive control system can transmit a prefabricated component structure model to the three-dimensional stereoscopic prefabricated component finished product appearance quality detection subsystem. The component transfer subsystem, upon receiving the transfer order, advances to a warehouse for ready loading of the prefabricated components. The aerial lift system hoists the required components to the quality detection points according to the received transfer order information, then the three-dimensional visual prefabricated component finished product appearance quality detection subsystem starts to detect whether the components are matched with the component information in the received order and whether the components have crack damage and other problems under the condition that the component transfer tool and the components reach the quality detection range, and then the component quality detection information is transmitted to the component comprehensive management and control system. The components qualified in quality inspection are hoisted to the component transfer tool by the hoisting system, the component-loaded transfer tool is transferred to a target area, and the transfer system can also feed back the working state and the position information of the transport tool to the component comprehensive management and control system in real time. The component comprehensive management and control subsystem feeds the type number, order completion condition and inventory information of the prefabricated components back to the mobile terminal.
And 2, after receiving a working starting instruction of the comprehensive component management and control subsystem, the three-dimensional stereoscopic vision prefabricated component quality inspection system firstly performs self-inspection on the system to determine whether the system can work normally, if the system cannot work normally, the system transmits an abnormal working instruction of the quality inspection system to the comprehensive component management and control subsystem, and the comprehensive component management and control subsystem feeds information back to the mobile terminal in time and sends an alarm prompt.
And 3, if the system self-inspection is normal, the protective cover is opened by the subaerial pre-embedded binocular vision camera below the quality inspection system control component quality inspection point so as to ensure that the binocular vision camera can normally acquire the three-dimensional image information of the appearance of the prefabricated component, and the protective cover is in a closed state when the inspection work is not carried out at ordinary times so as to protect the subaerial pre-embedded binocular vision camera. The quality inspection system then receives a component transfer order from the component integrated control subsystem.
And 4, monitoring whether the component transferring tool reaches a detection point by a transportation tool parking sensing module in the quality inspection system. If the component transportation tool does not arrive within the specified waiting time, the quality inspection system can continuously detect whether the transportation tool arrives within the polling detection time which is not more than the set waiting time, if the transportation tool does not arrive all the time, the quality inspection system enters a dormant state, and transmits a dormant information instruction to the component comprehensive management and control subsystem, the component comprehensive management and control subsystem can feed information back to the mobile terminal in time and send a dormant information prompt, and meanwhile, the component comprehensive management and control subsystem can also send an instruction to the component transportation subsystem to acquire the state of the component transportation subsystem.
And 5, if the component transfer tool arrives within the specified time, the component space sensing module in the quality inspection system immediately starts working to detect whether the prefabricated component reaches the detection range, if the prefabricated component does not arrive within the specified waiting time, the quality inspection system can also continuously detect whether the component arrives within the polling detection time which is not more than the set waiting time, if the prefabricated component does not arrive all the time, the system enters a dormant state and simultaneously transmits a dormant information instruction to the component comprehensive control subsystem, the component comprehensive control subsystem can timely feed information back to the mobile terminal and send a dormant information prompt, and the component comprehensive control subsystem can also send an instruction to the storage gantry crane system to acquire information whether the component is hoisted by the gantry crane system.
And 6, if the component is determined to reach the detection range, controlling an image acquisition module to start working by a master control module in the quality detection system, namely, starting working by an upper binocular vision camera and a lower binocular vision camera at the component quality detection point, and acquiring the appearance three-dimensional image information of the prefabricated component, correcting the cameras and acquiring the upper and lower layer disparity maps of the prefabricated component. After the image information is acquired, a master control module in the quality inspection system controls an image processing module to start working, the prefabricated part appearance three-dimensional image information acquired by the image acquisition module is matched and detected with a prefabricated part structure model (the prefabricated part model is transmitted to the image processing module of the quality inspection system from a superior management system, namely a part comprehensive management and control subsystem) and a matching error instruction is transmitted to the part comprehensive management and control subsystem if the matching fails, the part comprehensive management and control subsystem can timely feed back the information to the mobile terminal and send an alarm prompt, then the quality inspection system sends an instruction for placing the part into a wrong part area to a navigation and hoisting system, and the navigation and hoisting system hoists the mistakenly taken part to the wrong part area.
And 7, if the component to be detected is successfully matched with the component structure model, the detected component is proved to be the component contained in the received transfer order, and the master control module in the quality inspection system controls the image processing module to carry out next component crack damage detection operation and detect whether the component has quality problems such as cracks, damages and the like. If the image processing module in the quality inspection system judges that the quality problems such as crack damage exist in the detected component, the quality inspection system can transmit an unqualified detection instruction to the component comprehensive control subsystem, the component comprehensive control subsystem can feed information back to the mobile terminal in time and send an alarm prompt, and then the main control module controls the network communication module to send an instruction for placing the component into a bad component area to the navigation hanging system, so that the component with the quality problem is placed in the bad component area by the navigation hanging system.
And 8, if the components are successfully matched and the quality inspection is qualified, the master control module controls the network communication module to send an instruction for hoisting the prefabricated components to the transfer tool to the navigation crane system, and then the navigation crane system hoists the detected qualified prefabricated components to the component transfer tool.
And 9, the master control module in the quality inspection system sends a corresponding instruction according to whether the components needing to be transported in the transportation order are hoisted or not. If the transfer order is not complete and hoisting continues, the system continues to detect if a new vehicle arrives, as the multiple components required for each transfer order may require two or even multiple component transfer tools to be co-transferred. And if the transfer order is finished, the quality inspection system finishes the quality inspection work. The component transferring tool can transport the loaded prefabricated components to a construction site.
When the quality detection system is used as a subsystem for component quality detection, in order to check the quality detection effect and compare the quality detection effect with the detection effect when the quality detection system is used as an independent system, the same transfer order is still drawn up (the quantity of components required by the order is larger than the sample volume), and the prefabricated components (including wrong components and bad components) with the quantity the same as that of the previous experiment are adopted for testing. The sample volumes were still 50, 100, 200 pieces, respectively. When the sample capacity is 50 pieces (32 pieces are order internals and fracture damage exists in 5 pieces, 18 pieces are not order internals), the accuracy of component matching is 100%, and the accuracy of fracture damage detection is 100%; when the sample capacity is 100 pieces (76 pieces are order internals and 9 pieces have fracture damage, and 24 pieces are not order internals), the matching accuracy is 99%, and the fracture damage detection accuracy is 100%; when the sample volume is 200 pieces (148 pieces are order internals and 23 pieces have fracture damage, 52 pieces are not order internals), the matching accuracy is 96%, and the fracture damage detection accuracy is 99%. According to the experimental result, although the matching accuracy and the crack damage detection accuracy are not greatly improved, the whole system is more perfect and more rigorous, and the quality inspection and transportation efficiency of the prefabricated part is greatly improved while the systems are matched with each other. The time required to test one component is still about 20 seconds.
When the three-dimensional stereoscopic prefabricated part quality inspection system operates independently as a complete system, three-dimensional reconstruction is carried out on the prefabricated part through an upper binocular vision camera and a lower binocular vision camera at quality inspection points, defect detection is carried out on the basis of three-dimensional point cloud data, size information of the prefabricated part is obtained through a high-quality parallax image, and therefore component model matching and component quality inspection are carried out. Through enterprise research, the time required by the traditional manual detection of one prefabricated part is about 2 minutes, and the time required by the detection of the prefabricated part is only about 20 seconds by relying on a three-dimensional visual prefabricated part quality detection technology in the transportation and transfer process of the prefabricated part, so that the detection efficiency is greatly improved. And at present, many enterprises perform component appearance detection by referring to design drawings through artificial vision, the matching accuracy is about 85%, the crack damage detection accuracy is only 73%, the component matching accuracy of the invention is more than 95%, and the crack damage detection accuracy is as high as 99%.
According to the invention, two groups of binocular vision cameras arranged at the quality detection points of the components are used for acquiring the three-dimensional image information of the appearance of the prefabricated components in real time, the structural model matching detection of the prefabricated components and the surface damage detection of the prefabricated components are combined in the image processing process of the prefabricated components to automatically sort the components, so that the quality inspection automation is realized, the quality inspection result can be reported and pushed to a superior management system, and managers can conveniently and timely acquire the quality information of the prefabricated components. The invention solves the problems that the quality inspection result of the prefabricated part is not standardized and normalized in the conventional machine vision inspection means and the traditional manual quality inspection mode, carries out comprehensive and convenient identification and appearance damage detection before the finished prefabricated part enters the link of logistics transportation each time, reduces the loss of time and cost caused by transferring wrong parts and parts with quality problems, reduces the waste of resources, reduces the occurrence of events seriously influencing the construction progress of the assembly type building to a certain extent, ensures the construction period and reduces the construction cost.
A prefabricated component finished product quality detection system based on three-dimensional stereoscopic vision and application thereof can provide an accurate and efficient quality inspection detection means, meanwhile, quality inspection information of components and the transportation state of the components are reported in time through close coupling with other systems, basis is provided for decision judgment of construction site engineering managers, and the whole component warehousing transportation system is more automatic and intelligent. The whole system is more perfect and stricter, and the quality inspection and transfer efficiency of the prefabricated parts is greatly improved while the systems are matched with each other. Powerful support is provided for constructing a complete assembly type building integrated platform, and the intelligent level of the quality inspection technology of the building prefabricated parts is improved.
Claims (6)
1. Prefabricated component finished product quality detecting system based on three-dimensional stereoscopic vision, including total control module, image acquisition module, image processing module, transport means perception module, component space perception module and network communication module, its characterized in that: the general control module is respectively connected with the image acquisition module, the image processing module, the transport means parking sensing module, the component space sensing module and the network communication module, the general control module is responsible for comprehensive control of all functional modules in the prefabricated component finished product appearance quality detection system based on three-dimensional vision, information interaction and command control of an integral hardware platform, the hardware part of the image acquisition module mainly comprises a binocular vision camera and a video transmission line of component quality detection points, the image processing module detects prefabricated component model matching and prefabricated component appearance damage, the transport means parking sensing module comprises a transport means parking sensing instrument, the component space sensing module comprises an infrared detector, and whether the prefabricated component to be quality tested reaches the component quality detection points is monitored.
2. The system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision according to the claim 1, wherein: the binocular vision cameras are divided into two groups, one group is arranged at the upper end of the component space sensing module, and the other group is arranged on the working ground.
3. The system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision as claimed in claim 1, wherein: the detection steps are as follows:
step 1, after a master control module receives a work starting instruction, a three-dimensional stereoscopic vision prefabricated part quality inspection system firstly performs self-inspection on the system under the control of the master control module to determine whether the system can work normally or not, if the system cannot work normally, the system transmits the abnormal work instruction through a network communication module, and the quality inspection system sends an alarm prompt;
step 2, if the system self-checking is normal, the master control module controls the binocular vision camera protective cover to be opened, namely the binocular vision camera pre-embedded on the ground below the component quality detection point opens the protective cover to ensure that the binocular vision camera can normally collect three-dimensional image information of the appearance of the prefabricated component, the protective cover is in a closed state when the detection work is not carried out at ordinary times to protect a group of binocular vision cameras pre-embedded on the ground, and then the quality inspection system can receive a component transfer order through the network communication module;
step 3, the transportation tool parking sensing module monitors whether the component transportation tool reaches a detection point, if the component transportation tool does not reach within the specified waiting time, the quality inspection system continuously detects whether the component transportation tool reaches within the polling detection time not greater than the set waiting time, and if the component transportation tool does not reach all the time, the quality inspection system enters a dormant state, and then transmits a dormant information instruction through the network communication module to send a dormant information prompt;
step 4, if the component transfer tool arrives within the specified time, the component space sensing module immediately starts working to detect whether the prefabricated component reaches the detection range, if the prefabricated component does not arrive within the specified waiting time, the quality inspection system can also continuously detect whether the component arrives within the polling detection time which is not more than the set waiting time, if the prefabricated component does not arrive all the time, the system enters a sleep state, and meanwhile, a sleep information instruction is transmitted through the network communication module to send a sleep information prompt;
step 5, if the component is determined to reach the detection range, the master control module controls the image acquisition module to start working, namely, an upper group of binocular vision cameras and a lower group of binocular vision cameras of the component quality detection point start working, and acquisition of three-dimensional image information of the appearance of the prefabricated component, camera correction and acquisition of a parallax image of an upper layer and a lower layer of the prefabricated component are carried out; after the image information is acquired, the master control module controls the image processing module to start working, the three-dimensional image information of the appearance of the prefabricated part acquired by the image acquisition module is matched with a structural model of the prefabricated part for component model detection, if the matching fails, a matching error instruction is transmitted by the network communication module to send an alarm prompt, then the master control module controls the network communication module to send an instruction for placing the component into a wrong part area to the navigation and hoisting system, and the navigation and hoisting system hoists and transports the mistaken component to the wrong part area;
step 6, if the component to be detected is successfully matched with the component structure model, the detected component is proved to be a component contained in the received transfer order, the master control module controls the image processing module to carry out next component crack damage detection operation, and whether quality problems such as cracks, damages and the like occur to the component are detected;
if the image processing module judges that the quality problems such as crack damage and the like exist in the detected component, the quality inspection system transmits an unqualified detection instruction through the network communication module and sends out an alarm prompt, and then the master control module controls the network communication module to send out an instruction for placing the component into a bad component area to the navigation hanging system, so that the navigation hanging system hangs the component with the quality problem to the bad component area;
step 7, if the components are successfully matched and the quality inspection is qualified, the master control module controls the network communication module to send an instruction for hoisting the prefabricated components to a transfer tool to the gantry crane system, and then the gantry crane system hoists the qualified prefabricated components after detection to the component transfer tool;
step 8, the master control module sends a corresponding instruction according to whether all the components to be transported in the transportation order are hoisted;
and if the transfer order is not finished, continuing hoisting, and if the transfer order is finished, finishing the work by the quality inspection system because the plurality of components required by each transfer order can need two or even more component transfer tools to be jointly transferred.
4. The system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision according to the claim 1, wherein: the application of the system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision in the device comprises a transport tool (1), a control box (2), a transport tool parking perceptron (3), a part perception module (4) with infrared rays on two sides, a binocular vision camera (5), a suspension part (6), a wrong part area (7), a bad part area (8) and a part storage area (9), wherein a master control module, an image processing module and a network communication module are arranged in the control box (2), a transport means that is used for manual control whole quality inspection system work stops opening and stops, and transport means that sensing module was berthhed to transport means stops perception appearance (3) and both sides contain component sensing module (4) of infrared ray and detects transport means (1) and component and has arrived the quality testing scope, and two sets of binocular vision cameras (5) from top to bottom begin work, and this quality inspection system carries out the quality inspection work.
5. The system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision according to claim 4, wherein: the wrong workpiece area (7) is arranged behind the bad workpiece area (8).
6. The system for detecting the finished product quality of the prefabricated part based on the three-dimensional stereo vision according to claim 4, wherein: the application method of the prefabricated part finished product quality detection system based on three-dimensional stereo vision in the device comprises the following steps:
step 1, in a transfer order business processing process, a field manager knows the current construction progress requirement according to a construction schedule, a project planning table and other reference documents, then after a mobile terminal edits a transfer order, the order is sent to a component comprehensive control subsystem, the component comprehensive control subsystem can simultaneously send the transfer order to a component transfer subsystem and a three-dimensional stereoscopic prefabricated component finished product appearance quality detection subsystem, and simultaneously send a required prefabricated component structure model to the three-dimensional stereoscopic prefabricated component finished product appearance quality detection subsystem (the prefabricated component structure model is matched with component appearance three-dimensional image information acquired by a binocular vision camera so as to judge whether a detected component is an order internal component or not;
step 2, after receiving a working starting instruction of the comprehensive component management and control subsystem, the three-dimensional stereoscopic vision prefabricated component quality inspection system firstly carries out self-inspection of the system to determine whether the system can work normally or not, if the system cannot work normally, the system transmits a normal working instruction of the quality inspection system to the comprehensive component management and control subsystem, and the comprehensive component management and control subsystem feeds information back to the mobile terminal in time and sends an alarm prompt;
step 3, if the system self-inspection is normal, the quality inspection system controls a binocular vision camera pre-embedded on the ground below a component quality inspection point to open a protective cover so as to ensure that the binocular vision camera can normally acquire three-dimensional image information of the appearance of the prefabricated component, and when the inspection work is not carried out at ordinary times, the protective cover is in a closed state so as to protect a group of binocular vision cameras pre-embedded on the ground, and then the quality inspection system receives a component transfer order sent by a component comprehensive management and control subsystem;
step 4, a transportation tool parking sensing module in the quality inspection system monitors whether a component transportation tool reaches a detection point or not, if the component transportation tool does not reach within the specified waiting time, the quality inspection system continuously detects whether the component transportation tool reaches within the polling detection time which is not more than the set waiting time, if the component transportation tool does not reach all the time, the quality inspection system enters a dormant state and transmits a dormant information instruction to the component comprehensive control subsystem, the component comprehensive control subsystem timely feeds information back to the mobile terminal and sends a dormant information prompt, and meanwhile, the component comprehensive control subsystem also sends an instruction to the component transportation subsystem to acquire the state of the component transportation subsystem;
step 5, if the component transfer tool arrives within the specified time, a component space sensing module in the quality inspection system immediately starts working to detect whether the prefabricated component reaches a detection range, if the prefabricated component does not arrive within the specified waiting time, the quality inspection system can also detect whether the component arrives within the polling detection time which is not more than the set waiting time, if the prefabricated component does not arrive all the time, the system enters a dormant state and transmits a dormant information instruction to the component comprehensive control subsystem, the component comprehensive control subsystem can feed information back to the mobile terminal in time and send a dormant information prompt, and the component comprehensive control subsystem can also send an instruction to the storage gantry crane system to acquire whether the component is hoisted by the gantry crane system;
step 6, if the component is determined to reach the detection range, the master control module in the quality inspection system controls the image acquisition module to start working, namely, an upper binocular vision camera and a lower binocular vision camera of the component quality detection point start working to acquire the appearance three-dimensional image information of the prefabricated component, correct the cameras and acquire upper and lower layers of disparity maps of the prefabricated component, after the image information is acquired, the master control module in the quality inspection system controls the image processing module to start working, the appearance three-dimensional image information of the prefabricated component acquired by the image acquisition module and the structure model of the prefabricated component are subjected to component model matching detection, if matching fails, a matching error instruction is transmitted to the component comprehensive control subsystem, the component comprehensive control subsystem feeds the information back to the mobile terminal in time and sends an alarm prompt, and then the quality inspection system sends an instruction for placing the component into an error area to the aerial lift system, the aerial lifting system lifts and transports the wrongly-taken component to the wrong component area;
step 7, if the structural model of the component to be detected is successfully matched with the structural model of the component, the detected component is proved to be a component contained in the received transfer order, the master control module in the quality inspection system can control the image processing module to carry out next component crack damage detection operation, whether the component has quality problems such as cracks and damages or not is detected, if the image processing module in the quality inspection system judges that the detected component has quality problems such as damage cracks and the like, the quality inspection system can transmit a detection unqualified instruction to the component comprehensive control subsystem, the component comprehensive control subsystem can feed information back to the mobile terminal in time and send an alarm prompt, and then the master control module controls the network communication module to send an instruction for placing the component into a defective component area to the gantry crane system, so that the gantry crane system places the component with quality problem into the defective component area;
step 8, if the components are successfully matched and the quality inspection is qualified, the master control module controls the network communication module to send an instruction for hoisting the prefabricated components to a transfer tool to the navigation crane system, and then the navigation crane system hoists the qualified prefabricated components after detection to the component transfer tool;
and 9, the master control module in the quality inspection system sends a corresponding instruction according to whether the components to be transported in the transportation order are hoisted completely or not, if the transportation order is not completed, hoisting is continued, because a plurality of components required by each transportation order may need two or even a plurality of component transporting tools to be transported together, the system continues to detect whether a new transport tool arrives, if the transportation order is completed, the quality inspection system completes the quality inspection work, and the component transporting tools transport the loaded prefabricated components to a construction site.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110372277.7A CN113102269B (en) | 2021-04-07 | 2021-04-07 | Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110372277.7A CN113102269B (en) | 2021-04-07 | 2021-04-07 | Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113102269A true CN113102269A (en) | 2021-07-13 |
| CN113102269B CN113102269B (en) | 2022-08-19 |
Family
ID=76714252
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110372277.7A Active CN113102269B (en) | 2021-04-07 | 2021-04-07 | Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113102269B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114558794A (en) * | 2022-03-03 | 2022-05-31 | 南京苏胜天信息科技有限公司 | Machine vision artificial intelligence processing system and method thereof |
| CN114871126A (en) * | 2022-05-06 | 2022-08-09 | 苏州诚朴智能科技有限公司 | Shell size detection device |
| CN114943778A (en) * | 2022-07-26 | 2022-08-26 | 广州镭晨智能装备科技有限公司 | Reference plane determining method, detecting method, device, equipment and storage medium |
| CN115082842A (en) * | 2022-08-23 | 2022-09-20 | 创颖峻网络科技(深圳)有限公司 | E-commerce goods appearance verification and warehousing processing method and device |
| CN115100440A (en) * | 2022-08-22 | 2022-09-23 | 深圳市今朝时代股份有限公司 | Power supply supervision feedback system based on super capacitor energy storage |
| CN115880296A (en) * | 2023-02-28 | 2023-03-31 | 中国建筑第五工程局有限公司 | Machine vision-based prefabricated part quality detection method and device |
| CN116772729A (en) * | 2023-08-22 | 2023-09-19 | 中铁二十三局集团第一工程有限公司 | Method for detecting appearance size of bridge prefabricated part based on laser radar |
| CN117523367A (en) * | 2023-11-20 | 2024-02-06 | 南方电网调峰调频发电有限公司检修试验分公司 | Blockchain system for tool management |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11203588A (en) * | 1998-01-20 | 1999-07-30 | Denso Corp | Vehicle type discriminating device |
| US20160038980A1 (en) * | 2013-06-06 | 2016-02-11 | Wheel Recovery Systems, LLC | Core Wheel Processing System and Method |
| CN207013315U (en) * | 2017-07-13 | 2018-02-16 | 青岛思锐自动化工程有限公司 | The flange quality control system of view-based access control model detection |
| CN207095514U (en) * | 2017-07-31 | 2018-03-13 | 成都经纬时空科技有限公司 | Track plates on-line quick detection system |
| CN109264585A (en) * | 2018-10-31 | 2019-01-25 | 郑州桔槔智能科技有限公司 | Tower crane Unmanned Systems |
| CN111604281A (en) * | 2020-04-13 | 2020-09-01 | 天津中车机辆装备有限公司 | Part sorting method and system for rail transit |
| CN111842177A (en) * | 2020-06-12 | 2020-10-30 | 天津扬天科技有限公司 | Firebrick detection control method based on structured light vision |
-
2021
- 2021-04-07 CN CN202110372277.7A patent/CN113102269B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11203588A (en) * | 1998-01-20 | 1999-07-30 | Denso Corp | Vehicle type discriminating device |
| US20160038980A1 (en) * | 2013-06-06 | 2016-02-11 | Wheel Recovery Systems, LLC | Core Wheel Processing System and Method |
| CN207013315U (en) * | 2017-07-13 | 2018-02-16 | 青岛思锐自动化工程有限公司 | The flange quality control system of view-based access control model detection |
| CN207095514U (en) * | 2017-07-31 | 2018-03-13 | 成都经纬时空科技有限公司 | Track plates on-line quick detection system |
| CN109264585A (en) * | 2018-10-31 | 2019-01-25 | 郑州桔槔智能科技有限公司 | Tower crane Unmanned Systems |
| CN111604281A (en) * | 2020-04-13 | 2020-09-01 | 天津中车机辆装备有限公司 | Part sorting method and system for rail transit |
| CN111842177A (en) * | 2020-06-12 | 2020-10-30 | 天津扬天科技有限公司 | Firebrick detection control method based on structured light vision |
Non-Patent Citations (1)
| Title |
|---|
| 吴玉厚等: "基于改进引导滤波的预制构件吊点位置检测方法", 《沈阳建筑大学学报( 自然科学版)》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114558794A (en) * | 2022-03-03 | 2022-05-31 | 南京苏胜天信息科技有限公司 | Machine vision artificial intelligence processing system and method thereof |
| CN114871126A (en) * | 2022-05-06 | 2022-08-09 | 苏州诚朴智能科技有限公司 | Shell size detection device |
| CN114943778A (en) * | 2022-07-26 | 2022-08-26 | 广州镭晨智能装备科技有限公司 | Reference plane determining method, detecting method, device, equipment and storage medium |
| CN114943778B (en) * | 2022-07-26 | 2023-01-13 | 广州镭晨智能装备科技有限公司 | Reference plane determining method, detecting method, device, equipment and storage medium |
| CN115100440A (en) * | 2022-08-22 | 2022-09-23 | 深圳市今朝时代股份有限公司 | Power supply supervision feedback system based on super capacitor energy storage |
| CN115082842A (en) * | 2022-08-23 | 2022-09-20 | 创颖峻网络科技(深圳)有限公司 | E-commerce goods appearance verification and warehousing processing method and device |
| CN115880296A (en) * | 2023-02-28 | 2023-03-31 | 中国建筑第五工程局有限公司 | Machine vision-based prefabricated part quality detection method and device |
| CN116772729A (en) * | 2023-08-22 | 2023-09-19 | 中铁二十三局集团第一工程有限公司 | Method for detecting appearance size of bridge prefabricated part based on laser radar |
| CN116772729B (en) * | 2023-08-22 | 2024-01-09 | 中铁二十三局集团第一工程有限公司 | Method for detecting appearance size of bridge prefabricated part based on laser radar |
| CN117523367A (en) * | 2023-11-20 | 2024-02-06 | 南方电网调峰调频发电有限公司检修试验分公司 | Blockchain system for tool management |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113102269B (en) | 2022-08-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113102269B (en) | Prefabricated part finished product quality detection system based on three-dimensional stereoscopic vision and application thereof | |
| CN110202388B (en) | Automatic production line and production method for small-size parts | |
| CN106809730A (en) | The container automatic butt tackling system and hoisting method of a kind of view-based access control model | |
| EP3960692A1 (en) | Intelligent forklift and container position and posture deviation detection method | |
| WO2009052854A1 (en) | Device, method and system for recording inspection data about a freight container | |
| CN105480864A (en) | Automatic detecting and calibrating system and method for container crane | |
| CN115311592B (en) | Construction site material safety evaluation system based on computer vision technology | |
| CN110135782A (en) | Cargo managing method, apparatus and system | |
| CN113361953B (en) | Customs port cargo pipe on-site robot inspection method and system | |
| CN110286614A (en) | A kind of dump car Unattended remote monitoring method and its discharge system | |
| CN113888007B (en) | Informationized management method and system for preventing aircraft hidden load | |
| CN105712206A (en) | Container spreader pose detection system and method | |
| WO2021062619A1 (en) | Container lock pin sorting method and apparatus, and device and storage medium | |
| CN113998595A (en) | Automatic anti-collision control system of unmanned crane | |
| CN114359479B (en) | Building information input modeling method based on BIM modular modeling | |
| CN115180512A (en) | Automatic loading and unloading method and system for container truck based on machine vision | |
| CN210763984U (en) | RFID-based intelligent container management system | |
| JP2009107736A (en) | Coil loading method and device | |
| CN111056197B (en) | Automatic container transferring method based on local positioning system | |
| CN110844426A (en) | Automatic sorting unmanned material room and synchronization method adopting RFID technology | |
| CN219636755U (en) | AI box number discernment and high accuracy positioner | |
| CN217591104U (en) | Real-time dynamic positioning device for assembled residential products | |
| CN114892662B (en) | Full-automatic hoisting method suitable for underground prefabricated structure | |
| CN110085336A (en) | System and method for detecting Nuclear Plant number and position | |
| CN207993841U (en) | A kind of automatic identification matrix of integrated circuit envelope survey automatic identification wafer box |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |